DESCRIPTION (Applicant's Abstract): Understanding the genetic and molecular bases of complex traits is a challenging task. Among all complex traits, those that define aspects of species differentiation hold a special place in evolutionary biology. Our aims are: I) Development of the recurrent selection/backcross (RS/B) method for mapping and cloning genes of complex traits - The method relies on continual selection for the desired phenotype (say, High) while backcrossing the selected products recurrently to lines of the opposite phenotype (Low). The final product should be lines whose genomes are derived from the Low parent, except at or near loci pertaining to the High phenotype. Whole genome scanning of marker genotypes should allow precise mapping of the majority of loci controlling the phenotype. We plan to develop this method using both computer simulations and analytical modeling. The method is also applicable to the analysis of the dynamics of nascent speciation, as well as the concept of species at the molecular level (Ting et al. 2000). II) Application of the method to Drosophila racial differentiation in sexual behaviors - The trait chosen is the strong sexual preference of D. melanogaster from Zimbabwe for their own type (Z type), over flies from other continents (M-type). Selection will be based on the fact that only Z-type males succeed in mating with pure Z females; selected males will then be backcrossed to M females. The experiment, after 40 generations, will create 60 replicate lines carrying Z-derived DNA at or near loci that determine Z-type male behavior; elsewhere, they will be M-like. Flies from Generation 20 and 40 will be genotyped over Chromosome III using DNA markers to be developed. We expect to define the total genetic architecture accurately and to map some of the major Z-maleness loci to within 30 kb. In the final phase, we will carry out association studies between 30 natural isolates and the candidate genes from each mapped region.